Evaluating impact of receptive field in Encoder-Decoder and U-Net models for Lane Detection Segmentation
Invalid Date
Classification - assign a label
Image classification — single label
Cat or Dog
Semantic Segmentation -> assign a label to each pixel
Output -> Mask
Matrix Dot Product
Sum of Element-Wise multiplication
Downsampling
Opposite of Convolution
Upsample
Expand
Convolution Layers in Encoder
Transposed Convolution Layer in Decoder
Double Convolution Layers
Transposed Convolution Layer in Expanding Path
Skip Connections (Concat) — Spatial Context
| Model | Parameters | Difference with UNet |
|---|---|---|
| CNN 3x3 | 3,139,587 | |
| UNet 3x3 | 31,037,763 | 27,898,176 |
| CNN 5x5 | 8,713,219 | |
| UNet 5x5 | 81,241,411 | 72,528,192 |
| CNN 7x7 | 17,073,667 | |
| UNet 7x7 | 156,546,883 | 139,473,216 |
PyTorch DataLoaders – Batch training
Epoch – 10
Loss – PyTorch CrossEntropy
| Model | Parameters | Difference with UNet |
|---|---|---|
| CNN 3x3 | 3,139,587 | |
| UNet 3x3 | 31,037,763 | 27,898,176 |
| CNN 5x5 | 8,713,219 | |
| UNet 5x5 | 81,241,411 | 72,528,192 |
| CNN 7x7 | 17,073,667 | |
| UNet 7x7 | 156,546,883 | 139,473,216 |
IoU
Dice Coefficient
CNN-5
UNet-3
| Model | Parameters | Difference with UNet |
|---|---|---|
| CNN 3x3 | 3,139,587 | |
| UNet 3x3 | 31,037,763 | 27,898,176 |
| CNN 5x5 | 8,713,219 | |
| UNet 5x5 | 81,241,411 | 72,528,192 |
| CNN 7x7 | 17,073,667 | |
| UNet 7x7 | 156,546,883 | 139,473,216 |
CNN-5 > UNet-3
The UNet-3 performs slightly better than the CNN-5
Dice score .885 > .85 and the IoU score of .805 > .755
CNN-5 is 75% smaller than UNet-3